1. Field of the Invention
The present invention relates to an image reading apparatus utilizing plural solid-state imaging devices.
2. Description of the Prior Art
In the image scanning with a solid-state imaging device there has been proposed the use of plural imaging devices in order to improve the resolving power of the image or to improve the transmission speed of the image signals.
An example of such image reading method is shown in FIG. 1, employing solid-state linear imaging devices A, B in combination with lenses 3, 4 for scanning a line in two blocks. The optical systems composed of said lenses project the image to be scanned onto photosensitive portions of said solid-state imaging devices. In such method, the optical systems should be so arranged that the scanning line becomes continuous at the boundary of said blocks.
In practice, however, it is difficult to exactly adjust the positions of two imaging devices according to the geometrical calculation of optical paths, and even a slight displacement of the imaging devices after the adjustment will result in an uncontinuous scanning line.
In order to avoid this drawback there has been proposed a method of so arranging plural solid-state imaging devices as that they read the image in a partially overlapping manner and eliminating the thus overlapping portion from the signals to compensate the positional aberrations of the imaging devices. FIG. 2 illustrates an example of such method, in which shown are areas A, B to be read respectively by the solid-state imaging devices; a test pattern 11 positioned outside the scanning area of the original in order to be read by said solid-state imaging devices A, B prior to the image scanning; and a reference marker 12 provided in said test pattern 11 and having a size of N pixels larger than the area overlappingly read by said solid-state imaging devices A and B. Prior to the image scanning the solid-state imaging devices A, B scan said test pattern 11 in a direction C to determine the number of bits NA until the reference marker 12 is read by the imaging device A and the number of bits NB until the end of the reference marker read by the imaging device B, and N-NB is calculated from said number NB. In the succeeding image scanning, the solid-state imaging device A performs the scanning from the 0-th bit to the (NA+N-NB)th bit while the imaging device B performs the scanning from the 0-th bit, thus eliminating the overlapping image information. However such method is still defective in requiring calculating function and involving complicated signal processing.